The tpl Promoter of Citrobacter freundii Is Activated by the TyrR Protein

JOURNAL OF BACTERIOLOGY,0021-9193/97/$04.0010

Sept. 1997, p. 5914–5921 Vol. 179, No. 18

Copyright © 1997, American Society for Microbiology

The tpl Promoter of Citrobacter freundii Is Activated by theTyrR Protein†

HONG QIU SMITH AND RONALD L. SOMERVILLE*

Department of Biochemistry, Purdue University, West Lafayette, Indiana 47907

Received 5 May 1997/Accepted 15 July 1997

The ability of microorganisms to degrade L-tyrosine to phenol, pyruvate, and ammonia is catalyzed by theinducible enzyme L-tyrosine phenol lyase (EC 4.1.99.2). To investigate possible mechanisms for how thesynthesis of this enzyme is regulated, a variety of biochemical and genetic procedures was used to analyzetranscription from the tpl promoter of Citrobacter freundii ATCC 29063 (C. braakii). By computer analysis of theregion upstream of the tpl structural gene, two segments of DNA bearing strong homology to the knownoperator targets of the TyrR protein of Escherichia coli were detected. A DNA fragment of 509 bp carrying theseoperator targets plus the presumptive tpl promoter was synthesized by PCR and used to construct a single-copytpl-lacZ reporter system. The formation of b-galactosidase in strains carrying this reporter system, which wasmeasured in E. coli strains of various genotypes, was strongly dependent on the presence of a functional TyrRprotein. In strains bearing deletions of the tyrR gene, the formation of b-galactosidase was reduced by a factorof 10. Several mutationally altered forms of TyrR were deficient in their abilities to activate the tpl promoter.The pattern of loss of activation function was exactly parallel to the effects of the same tyrR mutations on themtr promoter, which is known to be activated by the TyrR protein. When cells carrying the tpl-lacZ reportersystem were grown on glycerol, the levels of b-galactosidase were 10- to 20-fold higher than those observed inglucose-grown cells. The effect was the same whether or not TyrR-mediated stimulation of the tpl promoter wasin effect. By deleting the cya gene, it was shown that the glycerol effect was attributable to stimulation of thetpl promoter by the cyclic AMP (cAMP)-cAMP reporter protein system. A presumptive binding site for thistranscription factor was detected just upstream of the 235 recognition hexamer of the tpl promoter. Thetranscriptional start point of the tpl promoter was determined by chemical procedures. The precise locationsof the TyrR binding sites, which were established by DNase I footprinting, agreed with the computer-predictedpositions of these regulatory sites. The two TyrR operators, which were centered at coordinates 2272.5 and2158.5 with respect to the transcriptional start point, were independently disabled by site-directed mutagen-esis. When the upstream operator was altered, activation was completely abolished. When the downstreamoperator was altered, there was a fourfold reduction in reporter enzyme levels. The tpl system presents anumber of intriguing features not previously encountered in TyrR-activated promoters. First among these isthe question of how the TyrR protein, bound to widely separated operators, activates the tpl promoter whichis also widely separated from the operators.

The ability of microorganisms to degrade L-tyrosine to phe-nol, pyruvate, and ammonia in a reaction catalyzed by theinducible enzyme L-tyrosine phenol lyase (EC 4.1.99.2) is re-stricted to certain genera within the Enterobacteriaceae family(23). Detailed studies, including the cloning and characteriza-tion of the structural gene for tyrosine phenol lyase, have beencarried out for Citrobacter freundii (3, 22), Erwinia herbicola(15, 39), and Escherichia intermedia (25).

In E. intermedia and E. herbicola, the induction of tyrosinephenol lyase is stimulated when cells are grown in the presenceof L-tyrosine (14, 23) and reduced in catabolite-repressed cells(14). Our current understanding (31) of the role of the TyrRprotein in controlling rates of gene expression in response toexogenous L-tyrosine in Escherichia coli made it seem likelythat the tyrosine phenol lyase (tpl) gene would belong to theTyrR regulon.

To investigate possible mechanisms for the regulation oftyrosine phenol lyase, we employed a variety of genetic andbiochemical procedures to analyze transcription from the tplpromoter of C. freundii. The results suggest that the tpl pro-

moter is subject to activation by the TyrR protein. In a parallelstudy (5), the tyrR gene of C. freundii has been characterizedand found to bear a high degree of sequence identity to thetyrR gene of E. coli. By several criteria, the tpl gene thusqualifies for membership in the TyrR regulon.

MATERIALS AND METHODS

Bacterial strains, bacteriophages, and plasmids. The biological materials usedare listed in Table 1. The P1 transduction method used to construct strainsSP1625 and SP1626 was similar to that described by Miller (30).

Media. Basal medium was salts mix E of Vogel and Bonner (41). The followingcompounds were included when appropriate: L-tyrosine, 50 mg/ml; 5-bromo-4-chloro-3-indolyl-b-D-galactoside (X-Gal), 40 mg/ml; glucose, 0.2%; glycerol,0.2%; and ampicillin, 50 mg/ml. All minimal media contained thiamine hydro-chloride (1 mg/ml) and biotin (0.1 mg/ml). Either L broth (26) or nutrient agar(31 g/liter; Difco) was used as complete medium for the propagation of bacterialstrains. Bacteriophage l derivatives were propagated and titrated on tryptoneagar (16).

DNA preparations. Plasmid DNA and M13 replicative-form DNA were puri-fied from cultures grown to saturation in L broth with Wizard Plus Miniprepresins purchased from Promega. Single-stranded M13 DNA was prepared by theprocedure of Sanger et al. (33). The preparation of competent cells and theirtransformation were carried out by the procedure of Mandel and Higa (27).When thermosensitive lysogens were being transformed, the heat step was omit-ted.

Chemicals and reagents. Restriction endonucleases, T4 DNA ligase, andDNA polymerase I large (Klenow) fragment were purchased from New EnglandBiolabs. E. coli RNA polymerase holoenzyme was purchased from Epicentre.TyrR protein was purified as previously described (11). TyrR protein concentra-

* Corresponding author. Phone: (765) 494-1614. Fax: (765) 494-7897.

† Journal Paper no. 15465 of the Purdue University AgriculturalResearch Station.

5914

on March 18, 2018 by guest

http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

TABLE 1. Bacterial strains, plasmids, viruses, and oligonucleotides

Strain, plasmid, phage,and oligonucleotide Description or sequence Source or reference

E. coli strainsDH5a F2 f 80 dlacDM15 D(lacZYA-argF)U169 deoR recA1 endA1 phoA

hsdR17(rK2 mK1) supE44 l2 thi-1 gyrA96 relA118

TG-1 supE thi-1 D(lac-proAB) D(mcrB-hsdSM)5 (rK2 mK2)[F9 traD36 proAB lacIq ZDM15]

Stratagene

NK5031 D(lacIZY) MS265 gyrA supF 29SP1312 F2 zah-735::Tn10 D(argF-lac)U169 19SP1313 F2 zah-735::Tn10 D(argF-lac)U169 DtyrR 19CSH26(lRZ11) ara D(lac-pro) thi (l plac5 cI857 Sam7) 457200 l2 e142 relA1 spoT1 D(cya-1400)::kan 35SP1625 SP1312(lHQS10) cya::kan This workSP1626 SP1312(lHQS10-mut1) cya::kan This work

Other bacteriaC. braakii (C. freundii) ATCC 29063

PlasmidspRVT tpl gene cloned in pBluescript 3pMLB1034 Promoterless lacZ; Ampr 6pJC100 tyrR1 gene cloned in pET3a 37pJC131 tyrR (D4) in pJC100 10pJC132 tyrR (D8) in pJC100 10pJC134 tyrR (L3A) in pJC100 10pJC135 tyrR (E4A) in pJC100 10pJC137 tyrR (R2K) in pJC100 10pJC138 tyrR (R2E) in pJC100 10pJC142 tyrR (L3K) in pJC100 10pJC144 tyrR (L3P) in pJC100 10pJC145 tyrR (L3W) in pJC100 10pJC146 tyrR (L3I) in pJC100 10pUC19 Cloning vector 45pUC19-tpl Wild-type tpl promoter in pUC19 This workpUC19-tpl D45 Deletion derivative of pUC19-tpl This workpUC19-tpl mut2D G3A change in Box B of the tpl promoter This workpUC19-tpl mut2D45 G3A change in Box B of the tpl promoter This workpUC19-tpl mut3 Changes to 235 hexamer of tpl promoter This workpHQS10 Wild-type tpl promoter in pMLB1034 This workpHQS10-mut1 tpl promoter in pMLB1034; G3A change in Box A This workpHQS10-mut2 tpl promoter in pMLB1034; G3A change in Box B This workpHQS10-mut3 tpl promoter in pMLB1034; altered 235 hexamer This workpHQS10-D65 Deletion derivative of tpl in pMLB1034 (Fig. 1) This workpHQS10-D130 Deletion derivative of tpl in pMLB1034 (Fig. 1) This workpHQS10-D179 Deletion derivative of tpl in pMLB1034 (Fig. 1) This work

Phagesm13mp18/p Promoter cloning vector 36m13mp18/p-tpl Mutagenesis substrate This workm13mp18/p-tpl-mut1 G3A mutation in Box A (see Fig. 1) This worklHQS10 tpl-lacZ reporter This worklHQS10-mut1 tpl-lacZ reporter with G3A mutation in Box A This worklHQS10-mut2 tpl-lacZ reporter with G3A mutation in Box B This worklHQS10-mut3 tpl-lacZ reporter with altered 235 hexamer This worklHQS10-D64 tpl-lacZ reporter with 64-nucleotide deletion This worklHQS10-D130 tpl-lacZ reporter with 130-nucleotide deletion This work

Oligonucleotide1355F CGGAATTCTAACTCACTGAAGCAAATAGCA1356F AGGGATCCGCCGGATAATTCATTTTGTTTC161H GTGTAAAGCAAAGTGTATAAAACGGCGTTG896H CTGAAAACTTAACTTTCTAGTCTGTGCTCA256J-II TGAGCACAGACTAGAAAGTTAAGTTTTCAG293J GCATCACCTACAACAGGCGCTTTTTAAATTATTGCCAG292J CTGGCAATAATTTAAAAAGCGCCTGTTGTAGGTGATGC1412I CGGAATTCTTTTAGTGATGT1413I CGGAATTCAGCAATACAGTA1275I CGGAATTCCCCTCTCCACG688I ATGTGGAGGCACTATGCAAGATTTAATTGA

VOL. 179, 1997 tpl PROMOTER ACTIVATION 5915


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

tions were determined spectrophotometrically at 280 nm, with an extinctioncoefficient of 34,470 M21 cm21 (42). a-35S-dATP and [a-32P]dATP were pur-chased from Amersham. Special-purpose oligonucleotides for use in PCR andsite-directed mutagenesis were synthesized in the Laboratory for Macromolec-ular Structure, Purdue University. All other chemicals were of the highest qualitythat was commercially available.

DNA sequence analysis. DNA sequences were determined by the methoddescribed by Sanger et al. (33), modified for the use of a-35S-dATP as thelabeling nucleotide (7). The reagents for sequencing were purchased from U.S.Biochemicals in kit form (Sequenase version 2.0) and used according to thedirections of the supplier.

Construction of l tpl-lacZ lysogens. tpl promoter fragments were inserted asEcoRI-BamHI fragments into pMLB1034. The resulting pMLB1034 derivativeswere transformed into CSH26(lRZ11). Lac1 l recombinants were isolated asdescribed by Yu and Reznikoff (45) with NK5031 as the plating indicator.High-titer lysates of Lac1 l recombinants were used to lysogenize strains SP1312(tyrR1) and SP1313 (DtyrR). Two independently isolated lysogens of each pro-moter construction were used in b-galactosidase assays.

Enzyme assays. b-Galactosidase was measured as described elsewhere (30).Cells were grown at 30°C to early log phase in Vogel-Bonner medium in thepresence of L-tyrosine (50 mg/ml) before being sampled. Assays were carried outin triplicate and have standard errors of #10%. The enzyme assay values arereported in Miller units (20).

DNase I footprinting. A DNA fragment of 527 bp carrying the tpl promoterwas released from pUC19-tpl by digestion with EcoRI plus PstI. The digest wassubjected to electrophoresis on 0.9% agarose. The promoter-bearing fragmentwas purified by adsorption to DEAE paper (Schleicher and Schuell) followed byelution with 20 mM Tris (pH 8.0) containing 0.1 mM EDTA and 1 M NaCl. TheDNA-containing solution was then phenol-chloroform treated and ethanol pre-cipitated. For use, the tpl promoter DNA was dissolved in water at a concentra-tion of 0.1 mg/ml. A sample of this fragment (0.3 mg) was selectively labeled atthe EcoRI site by treatment with DNA polymerase I (Klenow fragment) in thepresence of [a-32P]dATP according to the method of Brenowitz et al. (9). Theunincorporated radiolabel was removed with a G-25 spin column (BoehringerMannheim), and residual protein was removed with a QIA spin column (Qia-gen), according to the recommendations of the supplier. Radiolabeled DNA(15,000 cpm per tube) was allowed to interact at 37°C with TyrR protein and/orRNA polymerase holoenzyme (for a typical set of conditions, see Fig. 3) in 10mM Tris (pH 8.0)–5 mM MgCl2–1 mM CaCl2–100 mM KCl–2 mM dithiothreitolin the presence of bovine serum albumin (50 mg/ml) and sonicated calf thymusDNA (2 mg/ml). The final volume of each binding reaction mixture was 180 ml.After 30 min, each tube was treated with 5 ml of DNase I (0.5 mg/ml) for 2 min.Digestion by DNase I was halted by the addition of 40 ml of 50 mM EDTA (pH8.0). Each sample was then treated twice with 200 ml of phenol-chloroform-isoamyl alcohol (24:25:1). The DNA was precipitated with ethanol and resus-pended in 5 ml of loading dye, and the suspension was heated for 10 min at 85to 90°C and loaded onto an 8% acrylamide–6 M urea DNA sequencing gel. Afterelectrophoresis, the gel was exposed overnight at 270°C to Kodak XAR-5 film.

Preparation of RNA. RNA was prepared from 250 ml of mid-log-phase cells,according to the procedure of Salser et al. (32). Two sets of cellular RNA wereprepared, either from E. coli DH5a(pHQS10) or C. freundii ATCC 29063. TheE. coli strain was grown at 37°C in L broth plus ampicillin (50 mg/ml); the C.freundii strain was grown at 30°C in tryptic soy broth (Difco) supplemented with0.2% L-tyrosine. The RNA was stored at 220°C in the form of an ethanol slurry.A third preparation of RNA was obtained by in vitro transcription with pUC19-tpl as the template. The reaction mixture contained, in a final volume of 50 ml,2 mg of template, bovine serum albumin (10 ml of 250 mg/ml), RNA polymeraseholoenzyme (10 U), 2 mg of TyrR protein, 1 ml of RNase inhibitor, and ribonu-cleoside triphosphates (15 ml of 5 mM stock solution). After 30 min at 37°C thereaction was terminated with 50 ml of 40 mM EDTA containing 300-mg/mltRNA. The RNA was purified by two successive precipitations with ethanol andstored at 220°C.

Oligonucleotide-directed mutagenesis. The introduction of the G3A changein Box A (see Fig. 1) was conducted with M13mp18/p-tpl as template by using theSculptor in vitro mutagenesis kit from Amersham. The mutagenic oligonucleo-tide was 161 H (Table 1), and the procedure was according to the directions ofthe supplier. Changes to Box B and to the 235 hexamer of the tpl promoter weremade with the oligonucleotide pairs 896H–256J-2 and 292J-293J, respectively, byusing the Quick-Change Site-Directed Mutagenesis kit from Stratagene, accord-ing to the directions of the supplier. Each mutational change was verified byDNA sequencing and the removal or addition of a diagnostic restriction endo-nuclease cleavage site.

Primer extension analysis of transcriptional start point. A 10-ng sample of a30-residue oligodeoxynucleotide (688I [Table 1]), complementary to tpl mRNA(Fig. 1), was hybridized with 10 to 50 mg of RNA in 0.15 M KCl–0.01 M Tris (pH8.3)–1 mM EDTA by heating at 100°C for 2 min followed by quick cooling on ice.The extension reaction was carried out as described by Curtis (12) in a finalvolume of 15 ml for 40 min at 42°C. At that time, excess nonradioactive dATPwas added, and incubation was continued for an addition 40 min. The reactionwas stopped by adding an equal volume of sequencing stop solution and heatingto 90°C for 2 min. A 5-ml sample was loaded onto an 8% acrylamide–6 M urea

sequencing gel. The extension products were run next to a sequencing laddergenerated with the same oligodeoxynucleotide and pUC19-tpl as template.

RsaI protection assay for binding of TyrR protein to the tpl promoter. Tosimplify the interpretation of the protection assay, a derivative of pUC19-tpl withone less RsaI site was constructed. A sample of pUC19-tpl was digested withSnaBI and BamHI and then treated with the Klenow fragment of DNA poly-merase in the presence of an equimolar mixture of deoxyribonucleoside triphos-phates. The product of the latter reaction was circularized with T4 DNA ligase.This procedure eliminated the RsaI site at coordinate 1137 (see Fig. 1), gener-ating pUC19-tplD45. Reaction mixtures contained, in a final volume of 20 ml, 1.5to 2 mg of target DNA (pUC19 tpl D45), 1 mM Tris-Cl (pH 7.5), 1 mM MgCl2,5 mM NaCl, and 0.1 mM dithioerythritol. Digestion with RsaI (5 U/tube) wascarried out at 37°C for 2 h. The digestion products were analyzed by electro-phoresis on 0.9% agarose gels. When appropriate, TyrR protein, ATP, andL-tyrosine were also added. For specific conditions, see Fig. 5.

RESULTS

Analysis of DNA sequences upstream of tpl genes. The DNAsequences adjacent to the known tpl genes (3, 15, 22, 25, 39)were scrutinized for the presence of target sites for the TyrRprotein. A consensus table based on authentic TyrR boxes wasused as the reference system, as previously described (21).Several potential operator targets were detected in the regionupstream of the tpl gene of C. freundii, the sequence of whichwas determined by Antson et al. (3). With the first A of thestarting Met codon of the tpl structural gene designated coor-dinate 11, possible TyrR boxes were detected with centers atcoordinates 2434.5, 2321.5, 2205.5, 2116.5, and 247.5.

Construction of lacZ reporter systems specific for the tplpromoter. The tpl promoter was synthesized by PCR withpRVT (3) as the template and oligonucleotides 1355F and1356F (Table 1) as primers. Oligonucleotide 1355F hybridizedwith DNA sequences upstream of the putative TyrR box atcoordinate 2434.5. It was designed to introduce a cleavage sitefor EcoRI. Oligonucleotide 1356F was designed to permit anin-frame connection (via a BamHI site) between the first fivecodons of the tpl gene and the 9lacZ gene of pMLB1034 (6).The PCR product was digested with EcoRI and BamHI, yield-ing a DNA fragment of 509 bp that was inserted into similarlycleaved pMLB1034. The resulting plasmid was named pHQS10.The tpl-lacZ transcription-translation fusion of pHQS10 wasthen transferred by standard techniques (4) into an integra-tion-proficient derivative of bacteriophage l. The resultingphage was named lHQS10. Single-copy lysogens of the D(lac)host strains SP1312 (tyrR1) and SP1313 (DtyrR) were prepared.Double lysogens were identifiable on the basis of colony coloron X-Gal medium or through standard b-galactosidase assays.The enzyme data from these isolates were rejected. Essentiallyidentical procedures were followed in the construction of LacZreporter systems carrying mutationally altered tpl promoters(see below), except that mutagenesis was carried out prior toinsertion into pMLB1034. The mutagenesis substrates wereeither M13-tpl subclones or pUC19-tpl subclones. The se-quence of the tpl promoter is shown in Fig. 1.

Regulation by TyrR of the tpl promoter. Strains of E. colicarrying LacZ reporter systems specific for the tpl promoterwere grown and assayed for b-galactosidase activity (Table 2).In tyrR1 strains, there was substantial production of b-galac-tosidase. In strains bearing a deletion of the tyrR gene, therewas a reduction of approximately 10-fold in reporter enzymelevels. This effect was reversed in the tyrR deletion strain whena plasmid (pJC100 [37]) that mediated the expression of afunctional TyrR protein was introduced. These results sug-gested that the tpl promoter, like the mtr and tyrP promoters,was subject to positive control by the TyrR protein.

Mutationally altered forms of TyrR unable to activate tran-scription from the tpl promoter. In previous studies, it wasfound that a number of in-frame deletions and amino acid

5916 SMITH AND SOMERVILLE J. BACTERIOL.


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

switches near the N terminus of TyrR reduced or abolished thepositive control function of this protein. These structuralchanges did not diminish the ability of TyrR to negativelyregulate repressible promoters (10, 43). To compare the TyrR-mediated activation of tpl with the mtr system, a set of plasmidsspecifying mutationally altered forms of TyrR were introducedinto a D(tyrR) host strain carrying a tpl-lacZ reporter system.When the b-galactosidase levels of the plasmid-bearing strainswere measured (Fig. 2), there were variations in reporter en-zyme levels that closely paralleled the results that had beenobtained (10) with the activated mtr promoter. Mutationalalterations that abolished TyrR activation of mtr (D4, D8, L3K,L3P, and L3W) also disallowed activation of the tpl promoter.Lesions that partly reduced the activation of mtr (L3A, R2K,R2E, and L3W) had similar effects on the operation of the tplpromoter. Finally, certain changes in TyrR that did not alterthe activation of mtr (E4A and L3I) led to proteins that werefully functional in the activation of tpl. This set of results is fullyconsistent with a role for TyrR as an activator of the tpl pro-moter.

FIG. 1. The tpl promoter and associated regulatory elements. Only the nu-cleotide sequence of the messenger-equivalent strand is shown. The coordinatesystem has been changed from the original publication (3) by assigning 11 to thestart point of transcription. DNA fragments used in the functional analysis of thetpl promoter were synthesized by PCR, with pRVT1 (3) as the template. Cleav-age sites for restriction endonucleases EcoRI and BamHI were installed at theindicated locations (2329 and 1180) in order to facilitate the construction ofsingle-copy reporter systems. Cleavage sites for restriction endonuclease RsaI(2278 and 2198) that were used in studies of the binding of the TyrR protein toBox A operator targets (Fig. 3) are indicated. Residues marked by an asterisk(2280 and 2166) are the locations of operator mutations that were generatedduring the course of this study. Note that the change of G (2280) to A leads tothe elimination of an RsaI cleavage site and that the change of G (2166) to Acreates a cleavage site for BfaI. The 59 end points of three truncated derivativesof the tpl promoter (Table 1) are shown as broad arrowheads at coordinates2262, 2193, and 2144. The presumptive 210 and 235 recognition elements areunderlined with heavy lines. The presumptive target site for cAMP-CRP (coor-dinates 258 through 229) is outlined with dashed lines. The hybridization sitefor the oligodeoxynucleotide primer used in the determination of the transcrip-tional start point is indicated by double underlining.

TABLE 2. b-Galactosidase levels in strains harboring tpl-lacZ reporter systems

Strain Relevant genotype Carbon source b-Galactosidase(Miller U)

SP1312(lHQS10) tyrR1 Glucose 140SP1313(lHQS10) D(tyrR) Glucose 13.5SP1313(lHQS10)(pJC100) D(tyrR) tyrR1 Glucose 515SP1312(lHQS10 mut1) G3A in Box A Glucose 17SP1313(lHQS10 mut1) G3A in Box A Glucose 16SP1312(lHQS10 mut2) G3A in Box B Glucose 39SP1312(lHQS10) tyrR1 Glycerol 3,340SP1312(lHQS10 mut1) G3A in Box A Glycerol 375SP1625(lHQS10) tyrR1 D(cya) Glycerol 55.7SP1626(lHQS10 mut1) G3A in Box A Glycerol 9.25SP1625(lHQS10) tyrR1 D(cya) Glycerol 1 cAMP 219SP1626(lHQS10 mut1) G3A in Box A Glycerol 1 cAMP 12SP1312(lHQS10 mut3) Altered 235 region Glycerol 84

FIG. 2. Effects of single amino acid switches and deletion mutations withinthe TyrR protein on the ability to activate the tpl promoter. Into strainSP1313(lHQS10) were introduced a series of plasmids carrying different tyrRalleles (Table 1). The plasmid-bearing strains were grown in minimal salts-glucose medium and assayed as described in Materials and Methods. The 100%values used as the basis of comparison were 515 Miller U for assays involving thetpl reporter system and 5,090 Miller U for assays involving the mtr reportersystem. Dark bars, tpl reporter system of lHQS10; open bars, mtr reportersystem as described by Cui and Somerville (10).



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

Identification of TyrR binding sites by DNase I footprinting.A fragment of DNA identical to the segment that had beenused to construct the tpl reporter systems was preparativelyisolated from pUC19-tpl as described in Materials and Meth-ods. The fragment was selectively radiolabeled with 32P at theEcoRI end. The promoter-bearing fragment was mixed withTyrR protein and/or RNA polymerase holoenzyme under avariety of conditions and then treated briefly with pancreaticDNase. The resulting digest was analyzed by gel electrophore-sis (Fig. 3). Two regions of the tpl promoter were protected bybound TyrR protein. The first (designated Box A in Fig. 1) wascentered 434.5 nucleotides upstream of the tpl gene. The sec-ond (designated Box B in Fig. 1) was centered 321.5 nucleo-tides upstream of the structural gene. Both TyrR binding sites,identified earlier by computer analysis, were close matches tothe canonical TyrR target sequence (TGTAAAN6TTTACA).Protection of Boxes A and B was negligible in the absence ofRNA polymerase (data not shown), a feature of target recog-nition by TyrR that has not been previously encountered.

Site-directed mutagenesis of TyrR targets. The highly con-served first G residues of each TyrR target were changed to Aresidues, as described in Materials and Methods. The initialscreening for the desired changes was facilitated by the factthat the G-to-A change within Box A eliminates a RsaI site,while the G-to-A change in Box B creates a BfaI site. Thecomplete nucleotide sequence of each mutant promoter wasverified by DNA sequencing. Each mutant promoter was thenincorporated via pMLB1034 subclones into integration-profi-cient Lac1 derivatives of bacteriophage l for evaluation ofpromoter function in vivo. When the tpl promoter was alteredvia a G-to-A mutation in Box A, there was no detectableTyrR-mediated activation (Table 2). The reporter enzyme lev-els in TyrR1 cells carrying tpl promoters with Box A mutationswere essentially identical to those observed in D(tyrR) hosts,i.e., there was a 10-fold reduction in reporter enzyme levelscompared to that for the tpl promoter with an intact Box A.The effect of the G-to-A mutation in Box B was not as severe(Table 2). As measured in terms of differential rate of b-galactosidase synthesis, the Box B mutation reduced the TyrR-mediated activation of the tpl promoter by only 3- to 4-fold.

Deletion of TyrR target sites abolishes tpl promoter func-tion. To investigate whether the tpl promoter could functionwithout the participation of the TyrR protein, three alteredversions of the tpl promoter having progressively larger dele-tions of DNA from the upstream region were constructed. Thiswas accomplished by PCR with oligonucleotides 1412I, 1413I,and 1275I (Table 1) to shorten the tpl promoter by 65, 130, and179 nucleotide pairs, respectively. The endpoints of these de-letion derivatives are indicated in Fig. 1. The three truncatedtpl promoters were installed within lacZ reporter constructs inthe usual fashion. In no case was there significant b-galactosi-dase activity (data not shown).

Quantitative aspects of the binding of TyrR to Box A of thetpl promoter. To investigate TyrR-operator interaction in moredetail, a series of in vitro studies using a well-defined systemwas carried out. The protection of the cleavage site for RsaIwithin Box A (Fig. 1) formed the basis of a convenient assay forthe binding of TyrR. The binding of TyrR to Box A wasassociated with the appearance of a DNA fragment of 359 bp.Lack of operator protection was indicated by RsaI fragments of279 and 80 bp.

The RsaI protection assay (Fig. 4) confirmed the DNase Ifootprinting results and the site-directed mutagenesis studies.In the absence of ligands, the TyrR protein was able to formstable complexes with the Box A segment of pUC19-tplD45.Half-maximal protection was observed at a TyrR concentra-

tion of 15 pM. The affinity of the TyrR protein for Box A wasgreatly enhanced by ATP and tyrosine. In the presence of theseligands, half-maximal protection occurred at a protein concen-tration of 0.09 to 0.125 pM. Surprisingly, the protection byTyrR of the RsaI site within Box A of pUC19-tpl D45 did notrequire RNA polymerase. By DNase I footprinting TyrR-op-erator interaction had been undetectable in the absence ofRNA polymerase. The most likely reason for this difference isthat the two methods for observing TyrR-operator interaction

FIG. 3. DNase I footprinting analysis of TyrR binding sites within the tplpromoter. Incubations (final volume, 200 ml) were carried out in siliconizedmicrocentrifuge tubes according to the procedure of Brenowitz et al. (9) asdescribed in Materials and Methods. Left lane, (G1A) standards, prepared bytreatment with formic acid (28). A DNA fragment, labeled at one end, carryingthe tpl promoter was allowed to interact with proteins under several conditions.Lanes: 1, RNA polymerase holoenzyme (500 nM) plus TyrR (12 nM); 2, RNApolymerase holoenzyme (1,000 nM); 3, RNA polymerase holoenzyme (1,000nM) plus TyrR (12 nM). ATP (200 mM) and L-tyrosine (200 mM) were presentin each tube. The sequence shown reads from the 59 end (bottom of figure)toward the 39 end (top) as shown in Fig. 1.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

employed DNA fragments that were topologically different.When pUC19-tplD45 was treated with XmnI (a single site inpUC19 far removed from the tpl insert), the resulting linearDNA was unable to interact with unliganded TyrR protein, asjudged by RsaI protection assays. In the presence of ATP,TyrR protein was able to interact with Box A in linear DNA,but four to five times more protein was required for RsaIprotection compared to the situation when the target was cir-cular (data not shown). Evidently the circularity and/or state ofsupercoiling of pUC19-tpl enables TyrR to firmly associatewith its target in the absence of RNA polymerase. With linearDNA, TyrR-operator binding is either weak, or the complex, ifit does form, readily undergoes dissociation. When the BoxA-specific RsaI protection assay was used to study the bindingof TyrR to DNA molecules bearing lesions in Box B, therewere no observable differences from DNA molecules having awild-type Box B (data not shown).

Transcriptional start point of the tpl promoter. Three dif-ferent samples of RNA, prepared from C. freundii and from E.coli(pHQS10) and by in vitro transcription of pUC19-tpl wereanalyzed by primer extension with reverse transcriptase, asdescribed in Materials and Methods. In each case (Fig. 5), asingle transcriptional start point, corresponding to the A resi-due designated 11 in Fig. 1, was detected. This result showsthat tpl mRNA contains an untranslated leader region of 165nucleotides. Inspection of the region upstream of the transcrip-tional start point showed the presence of hexamers with sub-stantial homology to the 210 and 235 recognition elementstypical of s70 promoters (Fig. 1). Experimental support for thecorrectness of the 235 hexamer was obtained by site-directedmutagenesis (see below). The 210 hexamer was chosen on thebasis of its similarity to the consensus 210 hexamers reportedin other systems.

Role of the cAMP-CRP system in tpl promoter function.Suzuki et al. (40) recently reported that the formation of tplmRNA in E. herbicola AJ2985 was drastically curtailed whencells were cultivated in the presence of glucose. These workers

also detected a possible cyclic AMP (cAMP) receptor protein(CRP) target site just upstream of the 235 recognition ele-ment of the E. herbicola tpl promoter. Inspection of the tplpromoter of C. freundii revealed a similarly situated CRP site(Fig. 1). Several experiments were, therefore, carried out todetermine whether catabolite repression via the CRP systemwas a feature of the operation of the C. freundii promoter. TwoTyrR1 strains bearing tpl-lacZ reporter systems were culti-vated in medium containing the noncatabolite-repressing car-bon source glycerol. One of the two strains had a wild-type tplpromoter; the other bore a tpl promoter with a G3A changein Box A (Fig. 1) that prevented TyrR-mediated activation.The b-galactosidase levels in glycerol-grown cells (Table 2)were at least 20-fold higher than the enzyme levels seen inglucose-grown cells. The magnitude of the carbon source effectwas the same whether or not TyrR-mediated stimulation wasoccurring (Table 2 [compare rows 1 and 6 and rows 5 to 7]).This suggests that cAMP-CRP-mediated stimulation of tpl pro-moter activity operates independently of TyrR-mediated acti-vation. The overall amplitude of regulation of the tpl promotervia the two control mechanisms that have been demonstratedis thus about 200-fold.

Direct verification of a role for the cAMP-CRP system wasobtained by deleting the cya gene from the pair of strains usedin the previous experiment. This was accomplished by P1 trans-duction, with strain 7200 (35) as the donor. The resultingcya::kan derivatives (SP1625 and SP1626) were grown in glyc-erol-Casamino Acids medium either in the presence or theabsence of cAMP. In the absence of cAMP, the cya mutantstrains produced about 50 times less b-galactosidase than iso-genic cya1 strains (Table 2, rows 9 and 10). The production ofb-galactosidase was stimulated by a factor of 4 to 5 whencAMP (0.5 mM) was present in the growth media (Table 2,rows 10 and 11). The failure of cAMP to restore b-galactosi-dase levels to control values undoubtedly reflects the hydrolysisof this compound by cAMP phosphodiesterase.

FIG. 4. Binding of the TyrR protein to Box A in vitro. The interaction ofTyrR protein with pUC19-tplD45 and subsequent treatment with RsaI werecarried out as described in Materials and Methods. (A) No ATP or L-tyrosine.Lanes: 1, no TyrR protein; 2 to 8, TyrR protein present at final concentrations of300, 150, 60, 30, 15, 10, and 7.5 pM, respectively. (B) ATP (2 mM) and L-tyrosine(0.5 mM) present. Lanes: 1, no TyrR protein; 2 to 8, TyrR protein present at finalconcentrations of 30, 15, 7.5, 3.75, 2.5, 1.875, and 1.2 pM, respectively. The sizes(in base pairs) of fragments are indicated on the right. A DNA fragment of 80 bpwhich was released by RsaI cleavage within Box A (Fig. 1) was not visible in theseexperiments.

FIG. 5. Primer extension analysis of the tpl promoter. Lanes: 1, assay with anoligonucleotide primer corresponding to positions 119 through 148 of the tplsequence (Fig. 1) and total RNA extracted from C. freundii grown at 30°C intrypticase soy broth (BBL) supplemented with L-tyrosine (500 mg/ml); 2, totalRNA extracted from E. coli DH5a(pHQS10) grown at 37°C in L broth plusampicillin (25 mg/ml); 3, RNA prepared by in vitro transcription of pUC19-tplwith the s70 form of RNA polymerase. In all cases, annealing was accomplishedby heating for 2 min at 100°C followed by cooling in an ice bath. The extensionreactions were carried out at 42°C for 40 min with 120 U of reverse transcriptase.G, T, A, and C correspond to a parallel dideoxy sequencing ladder generatedwith the same primer. The sequence depicted on the left is the reverse comple-ment of the sequence read from the autoradiogram and corresponds to themessage-equivalent strand shown in Fig. 1.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

Site-directed mutagenesis of the tpl promoter. To confirmthe location of the tpl promoter, oligonucleotide-directed mu-tagenesis was used to change the 235 hexamer from TTTACAto ACAGGC. The resulting construct, named tpl mut 3, wasincorporated in standard fashion into a single-copy lacZ re-porter system. When cells harboring l HQS10-mut3 were as-sayed in a TyrR1 background, the b-galactosidase levels werereduced by more than 10-fold compared to control values(Table 2, row 12). This result suggests that the assignment ofthe location of the tpl promoter from the primer extensionassays is correct.

DISCUSSION

In this report, we have presented several lines of evidencethat transcription from the tpl promoter is subject to positivecontrol by the TyrR protein. In tester strains bearing a deletionof the tyrR gene, there was at least a 10-fold reduction in theproduction of b-galactosidase from a single-copy tpl-lacZ re-porter system. Promoter function was completely restoredwhen a plasmid expressing the tyrR1 gene was introduced intothis E. coli strain. When several mutationally altered tyrR genesbearing lesions known to alter the activation of mtr or tyrP wereseparately introduced into such tester systems, there were re-ductions in tpl promoter activity that exactly paralleled thosethat had been observed in these two other TyrR-stimulatedtranscriptional units. By computer analysis, three DNA seg-ments that might qualify as operator targets for TyrR wereidentified. Two TyrR boxes far upstream from the transcrip-tional start point (at 2272.5 and 2158.5) were shown to bevalid targets for the binding of TyrR by DNase I footprinting,by a restriction endonuclease protection assay, and by theintroduction of single nucleotide switches that either com-pletely or partially disabled the in vivo activation of the tplpromoter. The third computer-predicted operator target failedto bind TyrR in vitro but happened to lie in a region of the tplpromoter that was almost certainly a binding site for CRP (seebelow).

In contrast to the situation that prevails in the other TyrR-regulated promoters, the operator targets of the tpl promoterlie far upstream of the transcriptional start point and are sep-arated from each other by a considerable amount of DNA (114bp). In the case of the mtr promoter, there are two TyrRbinding sites whose centers are separated by 30 bp, just up-stream of the 235 recognition element for RNA polymerase(20, 34). The centers of the two TyrR boxes that are importantfor TyrR-mediated activation of tyrP are separated by 23 bp;one of the boxes overlaps the 235 region of the tyrP promoter(2). The importance to transcriptional activation alterations inspacing between TyrR boxes has been studied in the tyrP sys-tem by Andrews et al. (1). These workers found that TyrR-mediated activation became undetectable when the two TyrRboxes in the region upstream of the promoter were separatedby more than 32 bp. In the present study, we found that evenwhen pairs of TyrR binding sites were far upstream from thepromoter and separated by 10 helical turns of DNA, they stillcontributed in critical fashion to the control of transcriptioninitiation. When either or both TyrR boxes were eliminatedfrom the system, the transcriptional activity of the tpl promotereither became virtually undetectable (Box A) or was sharplyreduced (Box B).

If cooperative interaction between TyrR dimers is an impor-tant aspect of transcriptional activation, this process can evi-dently occur over distances much greater than those heretoforeobserved. Within the tpl promoter, the activation sites for TyrRlie far upstream of the RNA polymerase binding site, at coor-

dinates 2272.5 and 2158.5. The use of such locations, whichhave been classified “remote” by Gralla and Collado-Vides(17), suggests that looping out of intervening DNA could occurin order for RNA polymerase to be touched by the TyrRprotein. Whether such looping occurs, or whether accessoryfactors such as histonelike proteins play a role in the process,remains to be investigated.

Role of cAMP-CRP in transcription from the tpl promoter.The tyrosine phenol lyase system of C. freundii has a number offeatures in common with the tryptophanase system of E. coli.The two enzymes catalyze analogous pyridoxal phosphate-de-pendent a,b-elimination reactions. At the primary structurelevel, 43% of the amino acid residues are identical (3, 22). Bothtryptophanase and tyrosine phenol lyase are subject to catab-olite repression (8, 14). In the case of the tryptophanase pro-moter, catabolite repression is attributable to a perfect con-sensus cAMP-CRP binding site (TGTGAN6TCACA) centeredat coordinate 261.5 (13). A similar presumptive cAMP-CRPbinding site (TGTGA N6 TCACC) is centered at coordinate250.5 of the tpl promoter (Fig. 1). The rate of utilization of thetpl promoter increases about 10-fold when cells are grown onglycerol, a non-catabolite-repressible carbon source (Table 2).The introduction of a cya (adenylate cyclase) mutation essen-tially eliminated transcription from the tpl promoter in glycer-ol-grown cells. This effect was partially overcome by the addi-tion of cAMP. To a first approximation, the activation of the tplpromoter by cAMP-CRP and by the TyrR protein appears tobe independent and additive, rather than synergistic. As pre-viously noted (3), there does not appear to be a sequenceencoding a leader peptide between the tpl promoter and the tplgene. This is a major point of difference from the tryptopha-nase system, in which transcriptional antitermination mecha-nistically connected to the translation of a leader peptide is amajor determinant of gene expression (38).

Recently, the regulation of the tpl gene of E. herbicolaAJ2985 was analyzed by Suzuki et al. (40). Using Northernblotting methods, these authors demonstrated that tpl mRNAwas induced by tyrosine and repressed by glucose, in agree-ment with the general conclusions of the present study. Byprimer extension mapping, the start point of transcription in E.herbicola was found to lie 121 bp upstream of the initiationcodon of the tpl gene. In contrast, the untranslated leaderregion of C. freundii was 165 nucleotides long (Fig. 1 and 5).Suzuki et al. postulated the existence of a TyrR box centered atcoordinate 265.5, a CRP binding site at coordinate 240.5, anda target homologous to the lac operator centered at coordinate126 (40). This distribution of control elements is quite differ-ent from that of C. freundii (Fig. 1). Our analysis of the knowntpl promoters suggests that these punctuation elements are oftwo separate categories. The first is represented by C. freundii(3, 22) and E. intermedia (25), while the second is found withinE. herbicola (15, 39). Two features distinguish the two pro-moter classes. First, the distances between the transcriptionalstart points and the starting ATG of the structural gene aredifferent. Second, the locations of operator targets for theTyrR protein are dissimilar. Both classes of tpl promoter ap-pear to be subject to catabolite repression. In our scrutiny ofthe region upstream of the tpl promoter of E. herbicola, wedetected possible TyrR targets at approximately the same lo-cations as those that were identified in C. freundii. It would beof interest to investigate whether such sites are capable ofinteracting with the TyrR protein.

ACKNOWLEDGMENTS

For his advice and assistance during the early phases of this work, wethank Ihor Skrypka. We are also grateful to Robert S. Phillips for



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

supplying pRVT DNA and to the reviewers of this paper for theirhelpful comments.

Financial support for this work came from grants from the U.S.Public Health Service (GM22131) and the U.S. Army Research Office(DAAH 04-95-1-01 38).

REFERENCES

1. Andrews, A. E., B. Dickson, B. Lawley, C. Cobbett, and A. J. Pittard. 1991.Importance of the position of TyrR boxes for repression and activation of thetyrP and aroF gene in Escherichia coli. J. Bacteriol. 173:5079–5085.

2. Andrews, A. E., B. Lawley, and A. J. Pittard. 1991. Mutational analysis ofrepression and activation of the tyrP gene in Escherichia coli. J. Bacteriol.173:5068–5078.

3. Antson, A. A., T. V. Demidkina, P. Gollnick, Z. Dauter, R. L. Von Tersch, J.Long, S. N. Berezhnoy, R. S. Phillips, E. H. Harutyunyan, and K. S. Wilson.1993. Three-dimensional structure of tyrosine phenol-lyase. Biochemistry32:4195–4206.

4. Aronson, B. D., M. Levinthal, and R. L. Somerville. 1989. Activation of acryptic pathway for threonine metabolism via specific IS3-mediated alter-ation of promoter structure in Escherichia coli. J. Bacteriol. 171:5503–5511.

5. Bai, Q., and R. L. Somerville. Cloning and characterization of the tyrR genesof Citrobacter braakii and Salmonella typhimurium. Unpublished data.

6. Berman, M. L., and D. T. Jackson. 1984. Selection of lac gene fusions in vivo:ompR-lacZ fusions that define a functional domain of the OmpR geneproduct. J. Bacteriol. 159:750–756.

7. Biggin, M. D., T. J. Gibson, and G. F. Hong. 1983. Buffer gradient gels and35S label as an aid to rapid DNA sequence determination. Proc. Natl. Acad.Sci. USA 80:3963–3965.

8. Botsford, J. L., and R. D. DeMoss. 1971. Catabolite repression of trypto-phanase in Escherichia coli. J. Bacteriol. 105:303–312.

9. Brenowitz, M., D. Senear, E. Jamison, and D. Dalma-Weiszhausz. 1993.Quantitative DNase I footprinting, p. 1–43. In A. Revzin (ed.), Footprintingof nucleic acid-protein complexes. Academic Press, San Diego, Calif.

10. Cui, J., and R. L. Somerville. 1993. A mutational analysis of the structuralbasis for transcriptional activation and monomer-monomer interaction in theTyrR system of Escherichia coli K-12. J. Bacteriol. 175:1777–1784.

11. Cui, J., and R. L. Somerville. 1993. The TyrR protein of Escherichia coli.Analysis by limited proteolysis of domain structure and ligand-mediatedconformational changes. J. Biol. Chem. 268:5040–5047.

12. Curtis, S. E. 1987. Genes encoding the beta and epsilon subunits of theproton-translating ATPase from Anabaena sp. strain PCC 7120. J. Bacteriol.169:80–86.

13. Deeley, M. C., and C. Yanofsky. 1982. Transcription initiation at the trypto-phanase promoter of Escherichia coli K-12. J. Bacteriol. 151:942–951.

14. Enei, H., H. Matsui, H. Nakazawa, S. Okumura, and H. Yamada. 1973.Culture conditions for the preparation of cells containing high tyrosinephenol lyase activity. Agric. Biol. Chem. 37:485–492.

15. Foor, F., N. Morin, and K. A. Bostian. 1993. Production of L-dihydroxyphe-nylalanine in Escherichia coli with the tyrosine phenol-lyase gene clonedfrom Erwinia herbicola. Appl. Environ. Microbiol. 59:3070–3075.

16. Gottesman, M. E., and M. B. Yarmolinsky. 1968. Integration-negative mu-tants of bacteriophage l. J. Mol. Biol. 31:487–497.

17. Gralla, J. D., and J. Collado-Vides. 1996. Organization and function oftranscription regulatory elements, p. 1232–1245. In Escherichia coli and Sal-monella. Cellular and Molecular Biology ASM Press, Washington, D.C.

18. Hanahan, D. 1983. Studies on transformation of Escherichia coli with plas-mids. J. Mol. Biol. 166:557–580.

19. Heatwole, V., and R. L. Somerville. 1991. The tryptophan-specific permeasegene, mtr, is differentially regulated by the tryptophan and tyrosine repres-sors in Escherichia coli K-12. J. Bacteriol. 173:3601–3604.

20. Heatwole, V. M., and R. L. Somerville. 1991. Cloning, nucleotide sequenceand characterization of mtr, the structural gene for a tryptophan-specificpermease of Escherichia coli K-12. J. Bacteriol. 173:108–115.

21. Heatwole, V. M., and R. L. Somerville. 1992. Synergism between the Trprepressor and Tyr repressor in repression of the aroL promoter of Esche-

richia coli K-12. J. Bacteriol. 174:331–335.22. Iwamori, S., S. Yoshino, K.-I. Ishiwata and N. Makiguchi. 1991. Structure of

tyrosine phenol-lyase genes from Citrobacter freundii and structural compar-ison with tryptophanase from Escherichia coli. J. Ferment. Bioeng. 72:147–151.

23. Kumagai, H., H. Matsui, and H. Yamada. 1970. Formation of tyrosinephenol-lyase by bacteria. Agric. Biol. Chem. 34:1259–1261.

24. Kumaji, H., H. Yamada, H. Matsui, H. Ohkishi, and K. Ogata. 1970. Ty-rosine phenol lyase. J. Biol. Chem. 245:1767–1772.

25. Kurusu, Y., M. Fukushima, K. Kohama, M. Kobayashi, M. Terasawa, H.Kumagai, and H. Yukawa. 1991. Cloning and nucleotide sequencing of thetyrosine phenol lyase gene from Escherichia intermedia. Biotechnol. Lett.13:769–772.

26. Lennox, E. S. 1955. Transduction of linked genetic characters of the host bybacteriophage P1. Virology 1:190–206.

27. Mandel, M., and A. Higa. 1970. Calcium-dependent bacteriophage DNAinfection. J. Mol. Biol. 53:154–159.

28. Maxam, A., and W. Gilbert. 1980. Sequencing end-labeled DNA with base-specific chemical cleavages. Methods Enzymol. 65:499–560.

29. Meyer, B., R. Maurer, and M. Ptashne. 1980. Gene regulation at the rightoperator (OR) of bacteriophage l. II. OR1, OR2, and OR3: their roles inmediating the effects of repressor and cro. J. Mol. Biol. 139:163–194.

30. Miller, J. H. 1972. Experiments in molecular genetics, p. 352–355. ColdSpring Harbor Laboratory, Cold Spring Harbor, N.Y.

31. Pittard, A. J., and B. E. Davidson. 1991. TyrR protein of Escherichia coli andits role as repressor and activator. Mol. Microbiol. 5:1585–1592.

32. Salser, W., R. F. Gesteland, and A. Bolle. 1967. In vitro synthesis of bacte-riophage lysozyme. Nature 215:588–591.

33. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing withchain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463–5467.

34. Sarsero, J. P., and A. J. Pittard. 1991. Molecular analysis of the TyrRprotein-mediated activation of mtr gene expression in Escherichia coli K-12.J. Bacteriol. 173:7701–7704.

35. Shah, S., and A. Peterkofsky. 1991. Characterization and generation ofEscherichia coli adenylate cyclase mutants. J. Bacteriol. 173:3238–3242.

36. Somerville, R. L., B. T. Hagewood, B. D. Aronson, and V. M. Tesmer. 1992.Frameshifted derivatives and promoterless frameshifted derivations ofM13mp18 and M13mp19 useful in the mutational analysis of cloned genes orpunctuation elements. Methods Mol. Cell. Biol. 3:103–107.

37. Somerville, R. L., T.-L. N. Shieh, B. Hagewood, and J. Cui. 1991. Geneexpression from multicopy T7 promoter vectors proceeds at single copy ratesin the absence of T7 RNA polymerase. Biochem. Biophys. Res. Commun.181:1056–1062.

38. Stewart, V. J., and C. Yanofsky. 1985. Evidence for transcription antitermi-nation control of tryptophanase gene expression in Escherichia coli K-12. J.Bacteriol. 164:731–740.

39. Suzuki, H., K. Nishihara, N. Usui, H. Matsui, and H. Kamagai. 1993.Cloning and nucleotide sequence for Erwinia herbicola AJ2982 tyrosinephenol-lyase gene. J. Ferment. Bioeng. 75:145–148.

40. Suzuki, H., T. Katayama, K. Yamamoto, and H. Kumagai. 1995. Transcrip-tional regulation of tyrosine phenol-lyase gene of Erwinia herbicola AJ2985.Biosci. Biotechnol. Biochem. 59:2339–2341.

41. Vogel, H. J., and D. M. Bonner. 1956. Acetylornithinase of Escherichia coli:partial purification and some properties. J. Biol. Chem. 218:97–106.

42. Wilson, T. J., P. Maroudis, G. J. Howlett, and B. E. Davidson. 1994. Ligand-induced self-association of the Escherichia coli regulatory protein TyrR. J.Mol. Biol. 238:309–318.

43. Yang, J., H. Camakaris, and A. J. Pittard. 1996. Further genetic analysis ofthe activation function of the TyrR regulatory protein of Escherichia coli. J.Bacteriol. 178:1120–1125.

44. Yanisch-Perron, C., J. Vieira, and J. Messing. 1985. Improved M13 phagecloning vectors and host strains: nucleotide sequences of the M13mp18 andpUC19 vectors. Gene 33:103–119.

45. Yu, X. M., and W. S. Reznikoff. 1984. Deletion analysis of the CAP-cAMPbinding site of the Escherichia coli lactose promoter. Nucleic Acids Res.12:5449–5464.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

Documents

The tpl Promoter of Citrobacter freundii Is Activated by the TyrR Protein